ML Projects

Customer Churn Prediction

tabularchurnproduction

Predicting which customers are likely to churn for a SaaS platform.

Approach: Baseline: Logistic Regression → XGBoost → Feature selection → SHAP analysis.Key Metric: AUC: 0.89, Precision@Top10%: 0.72
DemoRepoDetails

NLP Support Ticket Routing

nlpclassificationbert

Automated routing of support tickets using NLP classification.

Approach: Baseline: TF-IDF + Logistic Regression → fine-tuned BERT → error analysis.Key Metric: Macro F1: 0.81, Routing accuracy: 92%
DemoRepoDetails

Energy Demand Forecasting

time-seriesforecastinglstm

Forecasting hourly energy demand for a utility provider.

Approach: Baseline: ARIMA → LSTM → Feature engineering → Hyperparameter tuning.Key Metric: MAE: 0.13, RMSE: 0.21
DemoRepoDetails

RAG-powered Knowledge Base

ragllmretrieval

Retrieval-Augmented Generation for internal knowledge base Q&A.

Approach: Baseline: BM25 → Dense retrieval → OpenAI GPT-3.5 → RAG pipeline.Key Metric: Top-1 accuracy: 84%, Mean reciprocal rank: 0.78
DemoRepoDetails